ngl I just found out about this website: https://evmobjectformat.org/
what an argument, we would hope so
I have been hearing similar observations and complaints from formalists and auditors for ten years now. Currently the worst case time for tracing the control flow of an EVM program is quadratic in the size of the program. That is impossibly slow at initialization time, and a huge obstacle offline. EOF fixes this.
3 Likes
Well, the EVM today is an ecosystem of many projects, most of which lack the resources or motivation to implement this. It’s unclear what the purpose is, since most people are satisfied with the current state of things.
Blockchains are meant to be immutable, with as few changes as possible—just look at Bitcoin.
The guiding principle should be making small changes with high impact. This, on the other hand, is a major change with minimal benefit.
I imagine what would Linux Torvalds say if Linux would make kernel API changes like this
I respectfully disagree with your definition of immutability. The principal of immutability applies to the ability to retroactively modifiy transaction history or contract bytecode, but not the rules of execution themselves. If a consensus can be reached about modifying EVM behavior without requiring that the existing state and chain history be modified then it should be implemented.
I further agree with the solidity team’s argument that it is too soon to seriously consider protocol ossification of the execution layer. While there is a debate to be had, it cannot be unanimously declared that Bitcoin’s decision to do such has been perfect to justify the same on Ethereum just yet.
1 Like
Hey Traub
I agree with you that Ethereum is not Bitcoin—because it needs to innovate. But there has to be a balance.
Bitcoin is a store of value because people can reasonably save for retirement, knowing that Bitcoin won’t change. Every time Ethereum changes, it undermines that same store-of-value narrative. Too much change over the past few years has significantly damaged ETH’s value relative to Bitcoin.
Moreover, changes come at a high cost to the Ethereum ecosystem, especially for smaller projects that are EVM-compatible. Most of these projects are stuck on much older versions of the EVM and Solidity.
Changes should originate from users. They should be voted on and implemented slowly and minimally. I’m not aware of any users who have asked for drastic changes to the EVM format.
If such requests do exist, they need to be quantified and documented. There must be a clear process for expressing different points of view and voting on them.
1 Like
-
While I don’t believe this forum is the correct place to discuss narratives about value capture, I disagree with the notion that change is solely a negative, especially when in many cases it can enhance the utility of Ethereum. It would not make sense to argue that EIP-1559 or the merge undermined the value of ETH.
-
Yes, concern about impacting smaller projects that are EVM-compatible is warranted, but the versioning schemes implemented by EOF do aid in resolving that by ensuring that future EVM upgrades are backwards compatible. I think the debate over
SELFDESTRUCTillustrates this well enough. -
If by users you mean validators then they can “vote” on this by choosing not to upgrade their nodes to latest versions. End-users who submit transactions don’t necessarily care about the format of the EVM itself. The majority don’t know what the EVM is, but they do care about new features for their transactions, and security, both of which EOF would have a positive second and third order effect on. If we’re going by what end-users specifically have asked for then most of the ETH roadmap would go away and we’d be spending all our time arguing about how to pump the price rather than making Ethereum better.
Hey Traub,
At the end of the day, users want Ethereum to be truly useful.
Would you agree that for most users, Ethereum is first and foremost a cryptocurrency and a store of value? They want it to have decentralized governance and a sound economic model — core principles that have always defined Ethereum’s appeal.
Are you suggesting that we should censor people on this message board simply for expressing these widely held views? If so, we would end up silencing 99% of Ethereum users — which hardly seems aligned with the spirit of decentralization and open dialogue.
Which specific new features are you referring to?
Let’s discuss them and figure out the minimum set of changes needed to implement this.
In most cases, when we speak of “users,” we mean paying customers. In the context of Ethereum, those paying customers are the people who send transactions. Would you agree? These users, in my understanding, want Ethereum to be fast, decentralized, and secure.
Stakers, on the other hand, deploy capital and are best thought of as investors. Their goals are different: they seek to see the value of the Ethereum token appreciate, and they expect staking rewards — which are derived from a combination of inflation and transaction fees.
Ultimately, the moral responsibility of Ethereum is to serve these two groups: users and stakers. The people employed by the Ethereum Foundation, as well as developers, researchers, validators, and others, are, in essence, paid to ensure that users and stakers are happy.
In nearly every corporate structure — for example, Microsoft — owning even a single share gives you the right to vote. The overwhelming majority of decentralized projects similarly have formal voting mechanisms. In fact, Ethereum is the only major project I am aware of that has no such formal governance system. In corporations like Microsoft, shareholder interests are protected by corporate law. In the decentralized world, where formal legal structures don’t exist, we rely instead on moral principles. And by those principles, if someone owns a portion of Ethereum, it seems reasonable — even necessary — that they should have a voice in its governance. Would you agree?
Vitalik Buterin himself has written extensively about quadratic voting, DAOs, and new forms of decentralized governance. Yet, when the community asks the Ethereum Foundation about governance, the response has been an uncomfortable and persistent silence — stretching over years.
This highlights another unfortunate reality: not only is it impossible to vote, but it’s also nearly impossible to even get a response to basic questions.
For example, if I buy a single share of Microsoft, I can attend a shareholder meeting and ask a question. In contrast, there’s no real mechanism for engaging with the Ethereum Foundation. Devcon sessions don’t allow for open Q&A. When submissions to Devcon are rejected, the submitters receive no feedback — not even a single line.
There’s also no straightforward way to report a security vulnerability to the Ethereum protocol, no structured incentivized testnets, and I’m not even talking about actually getting a bounty paid.
1 Like
Is it a store of value first and foremost? I would argue that Eth’s primarily value is derived from its utility in accessing decentralized computation, and if that is the case then EOF stands to enhance that by enabling the acceleration of ethereum’s base-layer development. I think that it is possible to have a sound economic model and still innovate, especially as EOF is a technical not economic upgrade.
If users want Ethereum to be useful, then we need to add new features, and the lack of things like versioning schemes or bytecode structure makes it more difficult to add those new features. I go back to selfdestruct because figuring out how to solve that issue is critical to implementation of the verge, which users do benefit from (more decentralization and security, lower validator requirements, etc).
I believe EIP’s such as EVMMAX are in the discussion as well as being a benefit of EOF which would enable new applications not currently feasible in the EVM (although I may be wrong about that my knowledge of the subject is limited).
I’m not suggesting that we censor anyone, you’re free to post whatever you want, I’m merely pointing out that this thread was intended for discussions of the technical aspects of EOF and politely ask that we stay on topic.